Trick or Treat
As a child, Halloween was incredible as you ran door-to-door and smothered yourself with sweets. But this had consequences – energy crashes, stomach aches, and if maintained at a sustained level, cavities. For many children, this is the first lesson that too much of anything can be a bad thing. An abundance of candy sounds incredible, but eventually the body needs its nutrients.

The software industry is beginning to learn these same lessons. Consumers have seen an abundance of apps, products, and companies swallow up the physical world. It began with search, expanded to video, spread to social, and enveloped retail. The most dominant companies in the world are software-based, and with this lens, it’s clear that software did eat the world.

But it didn’t eat all of it. In fact, the largest, juiciest parts of the world have barely been nibbled. We know they’re next on the menu. We know they’ll eventually be consumed. But first, we need to deal with the consequences of eating just candy.

Software ate (part of) the world. Now we're feeling our first sugar crash.

The Proclamation
Less than 10 years ago, Marc Andreessen made his “Software is Eating the World” proclamation. The world(’s financial markets) didn’t fully comprehend software’s growing influence, and Marc identified this disconnect. In the excerpt below from his WSJ article (emphasis mine), he enumerates why this shift to software was only beginning, and why software would be so powerful.

But too much of the debate is still around financial valuation, as opposed to the underlying intrinsic value of the best of Silicon Valley’s new companies. My own theory is that we are in the middle of a dramatic and broad technological and economic shift in which software companies are poised to take over large swathes of the economy. More and more major businesses and industries are being run on software and delivered as online services — from movies to agriculture to national defense. Many of the winners are Silicon Valley-style entrepreneurial technology companies that are invading and overturning established industry structures. Over the next 10 years, I expect many more industries to be disrupted by software, with new world-beating Silicon Valley companies doing the disruption in more cases than not.

In reflection, this prediction appears obvious. Software is everywhere! As of January, 2020, the five largest tech companies, which are (mainly) software-based, make up ~17.5% of the S&P 500. There are a record number of software engineer job openings across every industry. News coverage is saturated with tech. But, a priori, the world struggled to grok software. Just ask taxi drivers, hotel chains, and newspapers. People mistook software as a change in the tide, instead of seeing the tsunami that would reshape the entire beach. The sentiment has shifted: the world now understands that technology, and software specifically, dominate our world. There are exceptions of course, and software cannot replace every part of the world. I can’t speak for Marc, but I think this is implicit in his argument: it won’t become the world, but more and more companies need it to survive.

Few anticipated this shift, and Marc deserves credit for connecting the dots. However, this global transformation is incomplete. Software has more to eat. The uninhibited gorge has stopped however, since there are questions that need to be answered. The best way to identify the remaining courses on the menu is to understand what was already eaten.

Savoring the Snacks
Marc referenced a number of industries in the original manifesto, and the following six have largely been consumed.

1. Entertainment (Books, Music, Movies, Photos, Video Games)

2. Transportation (Cars, Hotels, Airplanes, Distribution)

3. Business Services (Cloud Storage + Compute, Recruiting, Teleconference)

4. Commerce (E-commerce, DTC, Marketing)

5. Energy

6. National Defense

This list it is not exhaustive, and as I’ll explain, the segments are still evolving. However, this list adequately illustrates the breadth of software’s reach. Like your ravenous uncle at the family tailgate, software ate these segments quickly, moving from one table to the next. To be clear, there are plenty of opportunities in these respective industries. New epochs take time to materialize, and software’s manifestation on the internet is no different. Netflix may have won the early battles of streaming, but competition has surged into the market. The rise of the streaming wars in 2019 only confirm we’re in nascent stages of such a transformation. This begets the question then, which of these six will change the most moving forward?

As I alluded to, there will be continued movement in the entertainment industry, especially in streaming, but aside from new launches like Quibi, I think this area will be rather stale in terms of wholesale transformation. Software changed the way we consumed content more than what the content actually is. Streaming has increased while cable subscriptions have fallen. Movies and shows will continue to move towards exclusive content on each respective platform. More content will be created for mobile first, but most people binge watch their shows on TV, so the mobile content might just be cast to a TV. Either way, mobile will continue to rise as the de facto consumption device.

In the next 10+ years, I anticipate we’ll see the largest accelerations in transportation, business services, and national defense. Developments of technology have often followed founders’ interests. This is fairly obvious; why would anyone start a company in a field they aren’t interested in! But it’s an important point because this has shaped what people have made. College students created social networks. Apartment dwellers in San Francisco rented out extra space to temporary guests. Stranded businessmen wondered why they had such trouble finding a taxi. The list goes on. The point is – many of these companies we created because the ideas they represented were obvious, and sexy. Managing supply chain operations isn’t sexy. So now that a lot of the sexy opportunities are gone, and software engineers are getting exposure to other industries, we’re going to see more transformation in less obvious places.

Transportation and business services would benefit from more software solutions. Supply chain management is obvious, but how will this change when fleets of delivery drones transform shipping lanes? Will cities create special traffic lanes for autonomous vehicles to deliver packages? These solutions will have to evolve as our transportation systems change. Or, take business services. There are multiple ways to have a video conference with a geographically dispersed team, but what about software for fully remote teams? Will we see analog digital offices where you can digitally see what your colleagues are doing without disturbing them? Will VR headsets eviscerate business travel? What cybersecurity firms will capitalize on increased remote work activity? Software can solve these problems, but as we dive into less obvious challenges, we’ll see more radical behavior changes.

Meanwhile, consider any software companies you know of in the energy and national defense industries. Of the six mentioned, these have been the least penetrated. Both require more physical assets and hardware, but software is starting to gain traction. Boeing’s recent failures show how crucial software programs are for the most advanced aerial technologies. While the 747 Max warrants an entirely separate discussion, it’s illuminating for the sake of this essay. As we push further on land, in the sea, and in the sky, we press further against boundaries imposed by nature. Computers are more accurate than humans, and they’re shouldering more of the load in our newer vehicles, devices, and weapons. Militaries have begun to pivot to electronic warfare, and drones exhibit one way that conflict is evolving. Eventually, wars will be fought in bits and bites — digital Trojan horse viruses attacking the cloud infrastructure of Troy.

As more of our conflict and economy shift to digital, energy demands will increase. Identifying ways to modernize energy infrastructure will require technology. Some of this will be with breakthrough science and hardware, but software can help control energy delivery. I sincerely hope that a grid shutdown or hack will be averted, but as the dependency on electricity rises, so too does the value in initiating a shutdown against enemies. This will drive rapid innovation and adoption in safeguards. Once one grid gets compromised, it will be obvious that immediate countermeasures must be taken. Countries that focus on military and energy software development will see tangible benefits for their investments. Software will need to protect our energy and military systems, but it also will need to protect everything else.

The six industries above have mainly been the beneficiaries of the recent wave of innovation, but they represent a meager amount of total US GDP. The remaining 70% is largely unpenetrated, and also happen to be some of the most exposed security-wise. These late-adopter industries are next on the menu.

New Foods, New Markets
Some of the most critical components of our world have barely changed during the internet era, and this means they’re some of the most vulnerable. And, since they interface with valuable, private information, they are also the most targeted. This is a dangerous recipe, and over the past five years we’ve seen record numbers of ransomware attacks on hospitals, colleges, and banks.

As fellow venture capitalist Keith Rabois has pointed out, most industries with relaxed regulations have seen a constant march toward more choice and lower price — you can certainly see it in computer hardware, for instance. But the industries where governments are most involved — health care, education and real estate — have seen dramatic price increases and constrained supply.

Predicting that these markets will be impacted by software isn’t by any means revelatory. That said, I believe the first half of this decade will show signs of initial progress, which will dramatically compound. By the end of the decade, healthcare will be patient-centric, education will be decentralized, and finance will be much smarter.

There is a lot of red tape, lock-in, and lobbying that have protected these three industries. There are reasons regulations exist though. It’s beneficial for society for drugs to be methodically tested before they’re approved. We don’t want waves of new banking apps that skirt FDIC regulations either. Laws preventing telemedine however, are antiquated, and we will have to eliminate and modify in order to serve the world as it is, not what it used to be.

Incentives structures also need to reworked. At the risk of oversimplifying each of these industries, this is what we ought to strive for:

• Healthcare systems for people to stay healthy

• Education built for people to learn skills they need to get jobs

• Finance systems for lower income populations to more easily grow wealth

There are other goals and objectives to each of these systems, but the point here is their main objective should be enhancing the lives of the people that use these services. People-centric systems will gain traction in this next decade.

The challenge to date has been that incentive structures and systematic obstacles have stalled innovation, and by extension, software’s entrance into these industries. A herculean force needs to unshackle software.

Unfortunately (though fortunately for software), these three industries have been plagued by immense inflation, and it’s only a matter of time before sweeping changes arrive. As to what will light the tinder box, I don’t know. I’m no augur. But it’s clear that public sentiment is shifting, and people are tired of paying for healthcare, education, and rent that exceed monthly paychecks.

Software still can eat a lot of the world - but these untapped industries will require an extra push. We’ll eventually see what courier delivers software its next meal.

New Flavors, New Geographies
There are 5.5 billion adults in the world, 5 billion have a phone, and 4 billion have smartphones. Yet, as of November 2019, 1.2 billion people in the world still didn’t have access to consistent electricity. Half of these people live in Africa. So on one hand, we’ve connected almost everyone, but on the other hand, sixteen percent of the world population doesn’t have access to an essential service! This is a huge opportunity for software, but the more interesting angle is which software companies will be adopted? The internet has largely been US-driven, but other countries are beginning to see global success. TicTok is a clear example, as are Spotify and Zoom.

Companies will emerge built on totally different assumptions than US ones. We're going to see a wave of startups that begin abroad, grow globally, and arrive in the US. Eventually, the US will be a consumer, and not the sole creator, of new technologies. The US will still innovate, but power can shift. Winners can turn into losers. Losers can turn into winners. We’re going to see cycles of innovation around the world.

It's like surfing. You just missed what looks like a big wave, and you sit in the water as your friends ride in towards the shore. But while they're riding their wave, you're waiting for the next one. You start paddling as the next one comes, and it turns out that this one is far bigger. You start surfing in and realize you're catching up to the other wave, and while that one has slowed down, yours is only speeding up.

China is riding that second, stronger wave now with QR codes and mobile payments. The US shot ahead early in payments with the credit card. But countries like China went from cash straight to QR codes, leapfrogging the rest of the world. The US may fall behind in some capacities, but its weaknesses will let it leapfrog forward when the time comes. Developing countries are similarly starting from sparse digital systems, and these modernize rapidly. The developed world will continue to plod ahead, but it will be leapfrogged in some areas, and this will lead to new waves of international startups.

Instead of having to cut all the red tape, imagine being able to build an entire healthcare system from scratch. You wouldn’t have to convert paper records to digital, because there aren’t any records to begin with. US assumptions aren’t relevant here, and local startups will seize their respective markets. These developing countries will have ample opportunities to do just that.

The US will not always dictate what’s on the menu, but that doesn’t mean we can’t enjoy what software has to offer. In the meantime, get ready for new cuisines and new popular dishes. Software’s cuisine is now global, but it might take some time for us to get used to these new flavors and spices.

Stomach Pains
Sometimes, when I eat really spicy food, I don’t feel well. We’re going to have similar experiences as software transitions away from a US-centric internet. Before software can continue eating, it needs to digest what it ate.

There’s a reason for this stomach ache, and its the speed in which software consumed industries, often without understanding the negative externalities. “Move fast and break things” is a great hacker mentality when you’re scaling up a social network, but it’s not appropriate for other use cases.

More of the world is expressed by software, and since software’s influence amplifies, that means that our problems have been pronounced by software. Tech analyst Benedict Evans recent presentation at Davos builds off this: we’re rediscovering old problems and falsely attributing them to technology. For example: lots of people are dismayed by Amazon’s growing influence in e-commerce (Benedict particularly likes when people ‘rediscover’ private label). Independent stores are going out of business, but the same thing happened when Walmart began.

There are some new problems society is dealing with, but they’re not usually the center of attention. Deepfakes now let people manipulate videos freely.

Deepfakes are to video what Photoshop was to images. We need new tools to verify authenticity. Drones currently are consumer entertainment devices, but what do we do when they disrupt an airport? Or, consider how connected we are now: you can find a group of like-minded people for any topic now. This is great for organizing a group on airport carpet enthusiasts, but it also means we have places online for people to advocate for drunk-driving, anorexia, and other heinous acts. The fringe of the town is no longer isolated, they now have a group that shares their twisted ideas, and this warrants an answer.

For each answer, there are trade-offs. Google and Facebook can afford to create moderation teams, but this means that they get to decide what goes on their platforms. This is fine when everyone agrees on what should be removed, but it gets much harder when you move towards more political topics.

Moderation isn’t binary. We can create tools to remove pornographic content, and we should certainly create tools to target child predators. What do we do though when people share something we don’t agree with? Political advertisements have been a hot topic, and I think that in the next few years we’ll see a divergence in how companies handle this. We’ll also see stark differences between countries; the United Kingdom has different speech laws than the United States, and they both dramatically differ compared to China. Now that software is more closely intertwined with the world, we’re going to see more rules and regulations that (hopefully) address the problems in an effective manner.

We’re dealing with new problems, and re-examining old ones. The printing press democratized the spread of information, and yet was lambasted for giving a voice to undeserving persons. The internet lets everyone be an author, and this means we get to hear what everyone in the world thinks. Often times, this isn’t necessarily desirable, but I’d say it’s safe to argue that the benefits outweigh the costs here. New technologies challenge us, but they move us forward as a species. Like a stomach ache, these too will pass. We can’t de-invent technology, but we can learn from it, change our behaviors, and prevent relapses.

Software’s digesting what it ate, which gives us time to confront technology’s externalities.

Nutrition Facts: The Food Groups
Would you eat something if it didn’t have a nutrition label? What if you were in Canada? What about India, or China?

The US created the Food and Drug Administration in order to protect the consumer, and we need new agencies for this digital age. We need new, informed conversations that acknowledge the world as it is, and not what it once was (or what some wish it were). Existing antitrust laws don’t work for this new digital economy.

Technology is amoral. It can be wielded for good and bad. Facebook connected the world, but we also connected all of the bad people. That isn’t a problem unique to Facebook, it’s a humanity problem. As software advances forward, more of these challenges will emerge. We can’t de-invent technology, which means that we’ll have to evaluate different trade-offs. These conversations will dominate the next 10 years as we try to regulate technology.

The best way to have an informed conversation, is to first become informed.

There’s another section of Marc’s proclamation worth revisiting:

Secondly, many people in the U.S. and around the world lack the education and skills required to participate in the great new companies coming out of the software revolution. This is a tragedy since every company I work with is absolutely starved for talent. Qualified software engineers, managers, marketers and salespeople in Silicon Valley can rack up dozens of high-paying, high-upside job offers any time they want, while national unemployment and underemployment is sky high. This problem is even worse than it looks because many workers in existing industries will be stranded on the wrong side of software-based disruption and may never be able to work in their fields again. There’s no way through this problem other than education, and we have a long way to go.

As a world, we’re horribly tech-illiterate. And in multiple ways too. Grandma doesn’t really know how to use her smartphone, and lots of people can’t change their default browser, but these are relatively easy to learn. The challenge, is that the more important technical knowledge isn’t widespread enough.

More and more companies are becoming dependent on software, but there also aren’t enough of them. This is a large labor gap. Unfortunately, even after 10 years, we still haven’t made strong strides to fill this. Not enough people can read and write software, let alone understand it. Software governs our lives in so many ways, and I find it troubling so few understand it. I hope the changes in education will facilitate some positive changes.

To be clear, knowing how to code isn’t essential to regulate tech companies or work at a software company. You do need a fundamental understanding of network effects, aggregation theory, and the other numerous frameworks that explain technology companies. Our academic and experiential understanding of this world of software is essential.

I dwell on this topic almost daily, and I will explore this deeply in future essays. The succinct summary: Marc is right again. We need to re-educate our students, workers, and regulators. Doing so will enhance earnings potential, accelerate job growth, and enhance the quality of life for not only the US, but the world.

We need better tools to understand the new world we live in, and right now, we don’t even have an ingredients list.

Bonn Ape-tit
Software’s already feasted on a portion of the world, but it’s not satisfied. In the 10 years following the Great Recession, Apple, Amazon, Alphabet, Microsoft, and Facebook entered the S&P 500 top 10. See their ascendancy below.

A lot has changed since August 2018. We’ve seen Apple, Microsoft, Alphabet (Google), and Amazon (albeit briefly), hit 1 trillion in total market cap by January 31, 2020. These companies aren’t exclusively software companies, but they represent the future of what companies will increasingly look like. They define success in the corporate world. They’re innovators, problem solvers, and the quintessential hybridization of man and machine. These characteristics provide a glimpse of what every other company will look like in the next 10, 20, & 50 years.

No company is without its own flaws, and tech has received a great deal of backlash in the past few years — some rightfully so. I disagree with many of the baseless attacks, but as I mentioned above, this largely stems from a lack of understanding these companies. Perception is reality, and even if some criticism is unfounded, it informs us of how fearful people are of these large, powerful companies. That however doesn’t change the fact that software is still consuming the world. It just reaffirms how resilient it is! Even in the face of opposition, it’s steadily grown in influence.

Software has really only just begun to change the world, and the next 10 years will be marked with continued seismic shifts.

Software ate the low-hanging fruit, now it’s ready for the rest of the meal.